[Mlir-commits] [mlir] [MLIR][AArch64] Refactor lowering of vector.contract to Neon I8MM (PR #149810)

[Andrzej Warzyński]

https://github.com/banach-space approved this pull request.

Thanks, LGTM!

Note to other reviewers - this was extracted from https://github.com/llvm/llvm-project/pull/148198, which is where I reviewed this change originally. Also, this update makes the NEON patterns more aligned with https://github.com/llvm/llvm-project/blob/main/mlir/lib/Dialect/ArmSVE/Transforms/LowerContractionToSVEI8MMPattern.cpp, so it's a step towards https://github.com/llvm/llvm-project/issues/145559.

Thanks @momchil-velikov , just one small nit inline.

https://github.com/llvm/llvm-project/pull/149810